We watch the news and some government politico (including Pres. Obama) talks about the 'recovery' and how many jobs have been created and saved and restored and . . . Usually the one speaking has no direct knowledge what he's talking about but relies on someone else for the data and frequently the speech being given.
This reminds me of when I kibbitzed on a speech 30 years ago.
My then wife was working - as administrator - with a Federally funded jobs program for the City of Roanoke; there was a 'consortium' which managed similar programs not including the City's. All workers in the program worked for local governments in some capacity as they were 'trained' with new skills.
Ex's program was managed properly - office staff minimally sized to do the job but no empire; the bulk of the funds actually went to the hired workers. The Consortium was not so well managed and ended up collapsing under its own fiscal mismanagement; the City of Roanoke (i.e. the Director of Finance and Ex) had to take over management, shut it down, and 150 workers in the area faced loss of their jobs .
Ex and I are at home, and she's got to prepare a speech for the DOF and press release about all the workers and where they would end up with the closing down. The good news is that a lot of the workers were picked up by their local government employer. Another bunch were able to use the skills learned to get a job in the 'private sector'. There were 'hard numbers' for the government hires; there was reasonable - but not exact - data for the private hires. Ex was pleased to show me her draft that had all 150 workers 'taken care of', and admitted some numbers came off the top of her head although based on her experience.
My response was that the report was not realistic. A couple of poor slobs wouldn't find anything - they'd be SOL (Straight Outta Luck). Ex agreed, but asked "well, how many". My response was that, politically, if you went over six percent it'd look bad, but under three percent would look phony. (Yeah, I was BS'ing at my best here, too). I think we settled on four percent.
Of course you can't write in a press release that four percent of the workers were SOL. So we went to the standby "other opportunities".
On TV news the next night, the City's Director of Finance (and acting administrator of the Consortium) was shown reading the press release Ex had written, and the media was eating it up. "55 percent were hired by the government they were assigned to in the program; 35 percent found private sector jobs, 6 percent resumed education courses (another discussion ex and I had) and 4 percent will pursue other opportunities".
I've never looked at government statistics the same way since.